Topik 6 Grafik pada R II

6.1 Interaktif Plot Menggunakan Plotly

Sekarang kita akan membahas beberapa metode manipulasi data untuk membersihkan dataset, menggabungkan berbagai dataset atau mengekstrak variabel dari bingkai data sebelum kita melompat ke beberapa dasar pemrograman.

library(ggplot2)
library(maps)
library(ggthemes)
library(plotly)
library(scales)
library(dplyr)
library(tidyr)

# download data
d1 = read.csv("https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_global.csv",
    check.names = FALSE)
# head(d1)

# rename Provice/State and Country columns
colnames(d1)[1:2] = c("State", "Country")

d1.2 = pivot_longer(d1, cols = -c(State, Country, Lat, Long), names_to = "Date",
    values_to = "Cases")
# convert dates

d1.2$Date = as.Date(d1.2$Date, format = "%m/%e/%y")

d2 = aggregate(d1.2$Cases, by = list(Lat = d1.2$Lat, Long = d1.2$Long, Country = d1.2$Country,
    Date = d1.2$Date), FUN = sum)
colnames(d2)[5] = "Cases"
# reorder
d2 = d2[, c(4, 1, 2, 3, 5)]

top1 = aggregate(d2$Cases, by = list(Date = d2$Date, Country = d2$Country), FUN = sum)
# select the last date to get overall total
top10 = top1[top1$Date == "2021-08-11", ]
# select top 10
top10 = top10[order(-top10$x), ][1:10, ]
# let's include Aus
top10_country = c(top10$Country, "Australia")

colnames(top1)[3] = "Cases"
data_p = top1[top1$Country %in% c(as.character(top10_country)), ]

p1 = ggplot(data = data_p, aes(Date, log(Cases), color = Country, group = Country)) +
    geom_line(stat = "identity", size = 1) + scale_x_date(labels = date_format("%m/%y"),
    breaks = "2 months") + theme_wsj()
p1

myCol2 = c("slateblue1", "purple3", "turquoise2", "skyblue", "steelblue", "blue2",
    "navyblue", "orange", "tomato", "coral2", "palevioletred", "violetred", "red2",
    "springgreen2", "yellowgreen", "palegreen4", "wheat2", "tan", "tan2", "tan3",
    "brown", "grey70", "grey50", "grey30")

p2 = ggplot(data_p, aes(Date, Cases, group = Country, color = Country)) + geom_line(size = 1.5) +
    geom_point(size = 1.5) + scale_colour_manual(values = myCol2, "Countries") +
    geom_text(data = data_p[data_p$Date == max(data_p$Date), ], aes(x = as.Date(max(data_p$Date) +
        4), label = Country), hjust = -0.01, nudge_y = 0.01, show.legend = FALSE) +
    expand_limits(x = as.Date(c(min(data_p$Date), max(data_p$Date) + 5))) + scale_x_date(breaks = seq(as.Date(min(data_p$Date)),
    as.Date(max(data_p$Date) + 5), by = "30 days"), date_labels = "%m/%y") + scale_y_continuous(labels = comma) +
    theme_classic() + theme(axis.title = element_text(size = 15, face = "bold"))
p2

fig_p2 = ggplotly(p2)
fig_p2

6.2 Plot Maps

Sekarang kita akan membahas beberapa metode manipulasi data untuk membersihkan dataset, menggabungkan

# take last day's data data from d2
d2 = d2[d2$Date == max(d2$Date), ]
world = map_data("world")
w1 = ggplot() + geom_polygon(data = world, aes(color = region, x = long, y = lat,
    group = group), fill = "white") + theme_map() + theme(legend.position = "none") +
    scale_fill_brewer(palette = "Blues")
map1 = w1 + geom_point(aes(x = Long, y = Lat, size = Cases, colour = Country), data = d2) +
    labs(title = paste("COVID-19 Cases as of ", as.character(unique(d2$Date))))
# static version
map1

# interactive version
map2 = ggplotly(map1, originalData = FALSE, tooltip = c("colour", "size"), width = 750)
map2

# To save htmlwidgets::saveWidget(map2,file='map2.html')

Kita akan menggunakan contoh dataset yang disebut us_stocks.csv
Mari kita impor menggunakan

read.csv